jurnal ict(computer-aided system for determining industrial)

8/18/2019 Jurnal Ict(Computer-Aided System for Determining Industrial)

1/14

175

Journal of ICT, 12, 2013, pp: 175 – 188

COMPUTER-AIDED SYSTEM FOR DETERMINING INDUSTRIALMACHINERY OPTIMAL REPLACEMENT PERIOD

Basil Oluwafemi Akinnuli 1 and Simeon Ayodele Babalola 2

Department of Mechanical Engineering Federal University of Technology

Nigeria

Corresponding author: [email protected] 1

ABSTRACT

This paper focuses on bridging the gap between age-dependentoptimal replacement theory and its industrial applications byidentifying and incorporating the criteria considered by eld(maintenance) engineers before embarking on industrial machinereplacement, into an existing model to make it appropriateindustrial-wise. We aim to economically quantify the identi edcriteria and thereafter integrate them into the weighted average

cost of running the machine from the time of its installationto the date of analysis to alleviate the drudgery in using thisapproach most especially when multiple machines are to beanalyzed, the Machine Replacement Analysis Software (MRAS)was developed and its implementation was done using dataobtained from a Grinder used for cocoa processing. A curve with

polynomial function and coef cient of determination of 0.935was obtained and this implies replacing the machine after theseventh year of usage.

Keywords: Industrial machine, maintenance, economic life, replacementmodel, criteria.

INTRODUCTION

The replacement of industrial machinery is a vital decision which has a

sumptuous reward if timely and accurately carried out. Replacement projectsare characteristically driven by technical obsolescence, requirements for

performance improvements or requirements for functionality improvementsince the functionality and performance of machines degrade as they approach


2/14

Journal of ICT, 12, 2013, pp: 175 – 188

176

the end of design life. Economically, excessive operating and maintenancecosts and higher reject/scrap rates are the factors often considered before

performing an age-dependent machine replacement.

Industrial machines, just like any other machinery experiences failures ata rate which is generally observed to be increasing as the machinery ages.Machinery failure can be de ned as any change in any of its parts or componentswhich cause the entire system not to be able to perform its intended functioneffectively (Bloch, & Geitner, 1999). When the frequency of failure becomeshigh, its effects can be catastrophic: injury, loss of moral, annoyance andinconveniencies to the operators. These may cause delays and erratic suppliesthat can affect customer goodwill, products standard which can affect thecompany’s image and lead to costly lawsuits in case of injury or loss of life.A machine due for replacement will generally lead to high loss of productionas a result of scraps/wastages, downtimes and so on. These are capable of

placing a company in a dangerous position of becoming uncompetitive.

The causes of failures can be due to faulty design, defects in its manufacturingmaterial, processing and manufacturing de ciencies, assembly or installationdefects, off-design or unintended service conditions, maintenance de ciencies(neglect, procedures) and improper or off-design operations (Bloch, & Geitner,

1999; Tadi ć , Vukeli ć & Jeremi ć , 2010). However, no matter how good asystem design, installation, and maintenance are, its parts or componentswears out with age.

considering the fact that machines can be classi ed as either a non-repairableor repairable system, an industrial machine was classi ed as repairable inthis model which means that they can be restored to satisfactory operation byany action, including parts replacements or changes to adjustable settings(NIST/SEMATECH, 2006). In industrial settings, the repair rate of themachinery has to be kept as low as possible to prevent loss in production andensure customer loyalty.

The complete life cycle of industrial machinery consists of its installation(after procurement), usage, the occurrence of failure at some time, the repairand restoration of the failed component, the occurrence of the second failureand its repair and restoration to service and so on. The failure and failure-free cycle repeats itself randomly (Tadi ć et al., 2010) until the machine is

retired from service. Despite the machinery’s life indicated by manufacturers,it is imperative for companies to determine the economic or optimal lifeof their machines which is principally in uenced by the manner of usageand maintenance. The most fundamental idea of any economic analysis of


3/14

177

Journal of ICT, 12, 2013, pp: 175 – 188

replacement is the difference between the physical and economic life of theequipment and there is a maintenance level for machines that either maximizes

pro t or minimizes cost (Collier, & Ledbetter, 1982). Replacement age for amachine that is placed on economic life arrives typically before fundamental

breakdowns result into worn-out and technological disabling (Khoub-bakht,Ahmadi, Akram, & Karimi, 2008). Every industrial machine has an economiclife that thereafter, using the machine is not economical.

Replacement decisions are critically important to a rm. The formulationof a replacement policy plays a major part in the determination of the basictechnological and economical progress of a rm as undue or hasty replacementcan be a serious drain on the operating capital that may be needed for other

bene cial uses, while postponing it beyond reasonable time will lead toincreasing production costs and less competitiveness on the part of the rm.

Although, development of the theories of replacement analysis had beencarried out over the past decades, their practical application is often neglected

by industrialists because of its inability to re ect some criteria fundamentallyconsidered by maintenance personnel (Tai, & Ching, 2005). In most theoreticalanalyses, the optimal replacement moment is expressed in relation to thecapital equipment’s age, whereas in empirical applications other factors also

play a part in timing replacements, as the output produced seems to be of highimportance (Bethuyne, 1998). It was observed through industrial visitationsembarked upon during the preliminary stage of this research that manyindustries in Nigeria depends on the personal judgment of their maintenanceengineers in making replacement decisions. This either connotes the problemof ineffective modeling of the industrial situation, or complexity duringapplication. this research work identi es the replacement criteria used bythese eld engineers, incorporates them into an existing model, and developA user-friendly software for easy application of the model.

METHODOLOGY

Combination of facts from industrial visitations and literatures reviewedwere used to identify the criteria which often triggers industrial machineryreplacement decision model building (which includes identi cation of the

input and output variables determination of the parameters and status variableand the logic of the model) design of documents for effective data collectionand the development of software to facilitate the easy execution of the modelin industries and the application of the model for generation.


4/14

Journal of ICT, 12, 2013, pp: 175 – 188

178

MODEL DEVELOPMENT

To determine the economical status of a machine for the purpose of replacementdecision, the time value of money must be considered by computing the

weighted average of all the costs of running the machine from its year ofinstallation till date as given by Sharma (2007):

(1)

Where A is the acquisition cost of the machine, R1 , R2, R3 …RT represents therunning cost of the machine at the end of the rst, second, till the T year; D represents the discounting factor; and denotes the running cost at year andrespectively and the expression between and in equation (1) represents theweighted average [which is denoted as in this paper] of all costs up to the

period with weights 1,,,… respectively. The given weights are actually thediscounted factors of the costs in the previous years.

Model assumptions

In this research work, it was assumed that a machine for replacement is

always available and during replacement, the machine is replaced with afunctionally comparable machine (that is, a machine which processes thesame raw materials and outputs the same products as the machine in use, butmay produce better quantity and quality). It was also assumed that equipmentutilization has a direct relationship with output unit.

Modi cations to the selected model

The integration of the criteria considered by industrialists is necessary to ensurethat the replacement decision encapsulates the vital industrial parameters foroptimal decision. Sharma (2007) only considered acquisition costs of machine(A), running cost of machine (R) and discounting factor (D) for his modeldevelopment but failed to take into account the yearly salvage value. Thismodel of Sharma can be de ned as the weighted average K(T), of all costsup to the period (T-1) with weights 1, D, D 2---- D T-1 respectively. Therefore,to take account of the yearly resale (or salvage) value of the machine, it iscustomary to deduct the discounted salvage value of the machine from its

acquisition cost. In addition to this, the operating cost needs to be adjusted inSharma’s model because operating cost is expected to increase gradually asthe machine ages. Although Sharma considered the value with reference tothe running cost of the preceding and subsequent year, it was observed that


5/14

179

Journal of ICT, 12, 2013, pp: 175 – 188

a curve with a minimum point at the replacement year can be establishedwhen values are plotted against its respective year and the curve has A highcorrelation with the general polynomial model which has been used by otherauthors (Khoub-bakht et al, 2008; Khoub-bakht, Ahmadi & Akram, 2010;

Of ong, 2001; Ogunlade & Akinbinu, 2004).

Therefore, THE equation (1) can be expressed as:

(2)

To take account of the yearly resale (or salvage) value of the machine withthe assumption that it is ripe for replacement, it is customary to deductthe discounted salvage value of the machine from its acquisition cost. TheEquation (2) can then be written as:

(3)

The salvage value is the estimated value that an asset will realise upon its saleat the end of its useful life and in situations where the machine’s parts aredissembled as spare parts; the salvage value is the summation of the worth of

the various parts of the machine when sold.

Identi cation of criteria that triggers machinery replacement decision

Through industrial visitations carried out at different manufacturing companiesin Nigeria, industrial machinery replacement criteria used in these industrieswere identi ed. From the results of the feedback received from interviewsconducted with maintenance engineers in the selected industries which werevisited, the following were identi ed as the critical factors that in uenced

industrial machinery replacement decision:

a) Excessive operating and maintenance cost – this leads to increase in the production cost which will eventually reduce the expected pro t marginor even lead to running the machine at a loss. To prevent this unpleasantincident, the machine is usually considered due for replacement(Akinnuli, 2009).

b) Reduction in output capacity of the machine – this can make itdif cult for the company to meet market demand, thereby denying thecompany the bene t of enjoying a larger market share. It can be caused

by the need for frequent readjustment, defective outputs and so on.Bethuyne (1998) also identi ed this factor as one of the most importantreplacement criteria.


6/14

Journal of ICT, 12, 2013, pp: 175 – 188

180

c) High frequency of failure – high failure rate has diverse negativeimpact on industries. For machine operators, it could cause lossof morale, injuries and even loss of life. It could also lead to loss ofcustomer goodwill. When maintenance level is not adequate to prevent

accumulation of tear and wear, frequency of failure will de nitelyincrease rapidly (Gupta & Hira, 2012).

d) High cost of penalties incurred as a result of material wastage, clean-ups, reprocessing and huge loss in production especially in processing

plants where the failure of a single machine can affect the whole production line and other unquanti able penalties such as loss ofcustomer loyalty.

e) Increase in downtime of the machine caused by longer repair time could

trigger the decision to replace such machine. A prolonged downtimecould also be due to absence of specialists who can repair such Amachine (Akinnuli, 2011).

f) Scarcity of spare parts which can make the machine irreparable(Akinnuli, 2009; Sharma, 2007).

Five cocoa processing industries were visited:

A- Coop- cocoa processing industry, Akure, Ondo State NigeriaB- Cocoa products Industry, Ede Osun State, NigeriaC- Stanmark cocoa company Ondo, Ondo State, NigeriaD- Tulib cocoa products, Sagamu, Ogun State, NigeriaE- Ile- Oluji cocoa products limited, Ile-Oluji Ondo State, Nigeria

The criteria which affected these companies are shown in Table 1 theinformation to justify this claims were collated from the machines job cards asthey affect excessive operating and maintenance cost, high frequency of failureand increase in downtime of machine caused by job cards. for reduction inoutput capacity of the machines, high cost of penalties incurred from materialwaste materials processing monitoring cards were used while for the scarcityof spare parts, spare parts inventory cards were used.

In the above generated table, it is deduced that scarcity of spare parts leads toreduction in output and capacity of machine increase in downtime of machine,while high frequency of failure leads to excessive operating and maintenance

cost, reduction in output and capacity of the machine, increase in downtime ofmachine and high cost of penalties incurred from material waste (which resultsfrom unload and reload of a machine, decoupling and coupling material owlines as well as leakages on the processing line).


7/14

181

Journal of ICT, 12, 2013, pp: 175 – 188

Table 1

Effect Criteria and Affected Company or Companies

Affected industriesS/No Criteria A B C D E F1 Excessive operating and maintenance cost × ×2 Reduction in output and capacity of the machine3 High frequency of failure due to machine age × ×4 High cost of penalties incurred from material waste × ×5 Increase in downtime of machine6 Scarcity of spare parts × × × ×

Computation for the running cost

Since it is desirable that industrial machinery replacement period be elongated,a good maintenance policy must be adopted. Separating the cost of scheduledand corrective (unplanned) maintenance will give the maintenance personnelthe opportunity to observe the effect of the present maintenance policy.The rate of occurrence of failure can also be monitored and this will givethe opportunity to know when to increase the frequency or the time spent on

scheduled maintenance per month so as to regulate the rate of occurrence offailure. The total running cost is a summation of the operating and maintenancecost of the machine.

(4)

Where is the running cost function of the machine at year , is the maintenancecost of the machine at year T, and is the adjusted operating cost at year T .

Computation for the adjusted operating cost

The monthly operating cost of the machine includes the cost of power or fuel,expendable accessories, lubricants and oil used in the month. The monthlyoperating cost is adjusted by multiplying it with the “Performance Ratio”of the machine to account for variability in the utilization intensity of themachine. The adjusted monthly operating cost is given by:

(5)

Where is the time in months, is the effective capacity of the machine, is theactual capacity of the machine and the ratio is termed the performance ratio ofthe machine. This ensures that the effect of the reduction in output capacity of


8/14

Journal of ICT, 12, 2013, pp: 175 – 188

182

the machine is factored into the model. The effective capacity of the machinemeans the rate of output of the machine given the product mix, schedulingdif culties and machine maintenance schedules while the actual capacityis the rate of output actually achieved after using the machine for a period.

Since the maintenance cost accounts for cost of lost times and penalties forthe downtimes, the performance ratio was used to adjust only the operatingcost of the machine to cater for situations which include, but not limited to,increase in setup time, need for more frequent lubrication, readjustments andrealignments, all of which affect the actual capacity of the machine.

The operating cost function is expected to increase gradually as the machineages. The adjusted yearly operating cost can be expressed as:

(6)

Computation for maintenance cost

It is imperative to know that the replacement policy will only be optimal whenall efforts had been taken to reduce the machinery running cost, especiallythe maintenance cost which in most cases increases greatly when machineryfailure occurs during production run. The monthly maintenance cost iscomputed as the summation of the planned or scheduled maintenance costand the unplanned (corrective) maintenance cost. The scheduled (or planned)maintenance cost is to cover the cost of materials, labor and cost per hourof planned downtime as a result of planned maintenance. Similarly, theunplanned maintenance cost is to cover the cost of materials and labor, cost

per hour of unplanned downtime (which caters for the loss in production) andwastages as a result of the unplanned maintenance.

Hence, the maintenance cost per month:

(7)

(8)

Where = planned maintenance cost per month, = corrective maintenancecost per month; = time of machine usage in months, = cost of materials

and labor of a planned maintenance, = Cost of + extra materials, labor and penalties due to failure, = cost per hour of planned downtime, = cost per hourof unplanned downtime; = downtime as a result of planned maintenance, =downtime as a result of maintenance forced by a failure.


9/14

183

Journal of ICT, 12, 2013, pp: 175 – 188

These monthly data are summed yearly to give the yearly maintenance costdata as follows:

(9)

Computations for the discounting factor

According to Of ong (2001), the discounting factor is a function of twovariables – interest rate and the in ation rate. Hence, the discounting factorcan be expressed as:

(10)

Integration of the identi ed criteria into the replacement model

The identi ed criteria have been expressed mathematically as shown inequations (4) to (9) and they can be easily integrated into equation (3) which

can be rewritten as:

(11)

Table 2

Summary of Model Modi cation

Initial Model Final Model

The objective is to nd the year that will be minimal as this will give theoptimal replacement period for the machine. Computing this value for a

single machine will require lots of data storage since a machine can beoperational for many years. This necessitates computer application, both inthe data management and mathematical analysis to make the model useful forindustrialists who need to concurrently manage much machinery.

Where: , and

K

Eqn (11)


10/14

Journal of ICT, 12, 2013, pp: 175 – 188

184

Software development

A software application was developed to facilitate the implementation of themodel in industries and to serve as a tool for the easy analysis of the machine

data. The software, Machine Replacement Analysis software, is also knownas MRAS 2011. It was developed using the Microsoft Visual Basic 2008

professional edition. The software basically consists of two parts – the databasesegment and the analysis segment as in Figure 1. The database segment allowsthe input of the data either from machine monthly or yearly utility cards. Everydata that is input is stored in the database and the hard copy of the data can

be printed out when necessary. The database also provides the data to be usedin the analysis segment. The analysis segment utilizes the data stored in thedatabase to compute the optimal replacement period using equation (11) anddisplays the optimum replacement period for the machine on a graph page.

Figure 1. Relationship between the utility cards and the software segments.

The replacement analysis result and graph, machine information, monthly andthe yearly report sheets are in printable formats. Figure 2 displays the menuof MRAS.

Figure 2. The menus of MRAS showing details of the le submenu.

MRAS

DATABASE ANALYSIS

MonthlyRecord Sheet

Yearly RecordSheet

Math Model

OptimalReplacement

Period

OUTPUTMonthly and Yearly Record Sheets,

Analysis/Graph

INPUT

Documents

*MonthlyUtility Card

OR

*YearlyUtility Card


11/14

185

Journal of ICT, 12, 2013, pp: 175 – 188

CASE STUDY

To illustrate the application of the model and its software, the data of anindustrial machine obtained at a cocoa processing company was used. The

yearly data used for the analysis are as shown in Table 3.

Table 3

MRAS Display of Yearly Data Used for the Analysis

Figure 3. Displays the report submenu, from which the results can be generated.

Figure 3. Display of the report submenu.

Analysis of results

Populating MRAS with the yearly data in Table 3, the replacement analysisresultS graph as processed and presented in the report menu of MRAS was

printed out as shown in Figure 4 and this was validated using the MS Access


12/14

Journal of ICT, 12, 2013, pp: 175 – 188

186

(See Figure 5). Performing regression analysis on the values of K(T) obtainedusing the MS Excel packages gave a polynomial trend with the equation andcoef cient of determination of 0.935 as shown in Figure 5.

Figure 5. MS access display of replacement analysis results.

From the results obtained, it can be seen that it is economical to replace themachine once it is in the 7 th year of usage because it gave the lowest weightedaverage cost []. After the 7 th year of running the machine, the production costof using the machine starts increasing which indicates that the machine is nomore economical to use.

Figure 6. MRAS displaying the number of machine failures per year(orange bar), amount spent on preventive maintenance (blue bar), correctivemaintenance (red bar), and scheduled maintenance (green bar).

K(T) = 2740 T2 - 39259T + 56619R² = 0.935

N 350,000.00

N 400,000.00

N 450,000.00

N 500,000.00

N 550,000.00

N 600,000.00

0 2 4 6 8 10 12

K ( T )

Year, T

K(T)

Poly. (K(T))


13/14

187

Journal of ICT, 12, 2013, pp: 175 – 188

For the cocoa grinder investigated in this case study, Figure 6 shows the bar chart on maintenance costs, operating costs and the number of machinefailures experienced per year.

CONCLUSION

It is worth stating that when industrial machinery has never required expensiverunning cost, the replacement decision based on the age of the machine neverarises. Other circumstances such as a need for change in product designand speci cations can also prompt a replacement decision. With a view toassist industrial engineers in identifying the optimal replacement periodfor industrial machineries, this research work itemizes the basic criteriaoften considered in industries, and a software package (known as MRAS

– Machinery Replacement Analysis Software ) was developed for the easyexecution of the model in industries. The software developed was capable of

producing the replacement analysis table, annual running cost chart, as well asthe replacement analysis graph in printable formats.

In summary, this research work has shown how management science can beemployed in the analysis of machine replacement decision in industries. The

impact of computer software in successfully managing and identifying theoptimal replacement period of industrial machinery cannot be overemphasized.

REFERENCES

Akinnuli, B.O. (2009). Development of models for machinery evaluation inmanufacturing industries (Unpublished doctoral dissertation).The Federal University of Technology, Department of MechanicalEngineering, Akure, Nigeria.

Akinnuli B.O (2011). Computer aided system for modeling machinery procurement due-date prediction in production industries. Journal of ICT , 10, 15.

Bethuyne, G. (1998). Optimal replacement under variable intensity ofutilization and technological progress. The Engineering Economist ,

43(2), 85-106.

Bloch, H. P., & Geitner, F.K. (1999). Machinery failure analysis andtroubleshooting . Gulf Professional Publishing.


14/14

Journal of ICT, 12, 2013, pp: 175 – 188

Collier, C. A., & Ledbetter, W.B. (1982). Engineering cost analysis . NewYork: Harper and Row Publishers.

Gupta, P.K. & Hira, D.S. (2012). Replacement models, operations research

(revised edition). New Delhi: Chand Company Ram Najar.

Khoub-bakht, G., Ahmadi, H., Akram A., & Karimi, M. (2008). Determinationof optimum life (economic life) for mf285 tractor: A case study incenter region of Iran American-Eurasian. Journal of Agricultural &

Environmental Science , 4 (1), 81-85.

Khoub-bakht, G., Ahmadi, H., & Akram, A. (2010). Determination of optimumlife (economic life) for mf285 tractor based on repair and maintenancecosts: A case study in center region of Iran American-Eurasian. Journalof Agricultural Technology , 6 (4), 673-686.

NIST/SEMATECH. (2006). Engineering statistics handbook. Retrieved fromhttp://www.itl.nist.gov/div898/handbook/

Of ong, A. (2001). Determination of optimal replacement period for a groupof machines. NJEM , 2(4), 74-79.

Ogunlade, B. J., & Akinbinu, A. F. (2004). A heuristic for old equipmentreplacement by small scale industries in an in ationary economy.

NJEM , 5 (1), 25-30.

Sharma, J. K. (2007). Operations research theory and applications (3 rd ed.).India: Macmillan.

Tadi ć , B., Vukeli ć , D., & Jeremi ć , B. (2010). De ning a theoretical model ofwear-caused failure of tool machine elements. Tribology in Industry ,32(3)

Tai , A.H., & Ching, W. (2005). On the use of renewal theory in machinereplacement models. International Journal of Applied MathematicalSciences , 2(2), 240-247.